The field of the disclosure relates generally to digital transmission systems, and more particularly, to wired, wireless, and optical digital transmission systems employing active carriers and amplification.
Conventional OFDM transmission technology has a high crest factor, which is the ratio of peak-to-average radio frequency (RF) power. Data Over Cable Service Interface Specification (DOCSIS) technology utilizes OFDM, and experiences a crest factor of approximately 16 decibels (dB). In present usage, cable operators implement one or more carriers of DOCSIS, or DOCSIS version 3.1, at high ends of the downstream cable bandwidth, and operate present coaxial amplifiers with a steep up-tilt, e.g., approximately 15 dB. DOCSIS 3.1 signals may be as wide as 190 megahertz (MHz), and wider bandwidth typically signify greater RF power. Thus, in this example, a DOCSIS 3.1 OFDM signal is considered to be of high RF power, which will particularly stress the dynamic range of an amplifier at the time a transported OFDM signal crests. Moreover, as an amplifier ages, its performance with respect to nonlinear distortion may deteriorate. This problem may be further compounded by the fact that nonlinear distortion in digital signals resembles, on conventional test equipment, like random noise, and is thus difficult to discern from other signal impairments, such as the random noise itself, or other interference.
Common path distortion (CPD) is one example of nonlinear distorter on a cable plant. CPD may be created, for example, by corrosion diodes, and if made by digital carriers, also resembles random noise on test equipment. CPD affects both upstream and downstream signals, and thus there is a need to be able to find and repair CPD problems, as well as a need to be able to find and repair distorting amplifiers that are out of balance, defective, or being improperly operated. In many instances, CPD occurs from connectors, which may or may not be associated with amplifiers. Accordingly, there is also a need to determine, where amplifiers are cascaded, which particular element(s) (e.g., which of several amplifiers) of the cascade are the source of the distortion.
Upstream DOCSIS 3.1 implementations, for example, in a cable modem transmitter, utilize orthogonal frequency division multiple access (OFDMA) modulation. Conventional cable plants may include hundreds of such transmitters operating in burst mode, making it difficult to detect and locate a defective cable modem transmitter on that node. In one example, where a single cable modem is utilizing a maximum bandwidth of 42 MHz, the single cable modem is prone to clip the upstream laser (or A-D converter) when its OFDMA signal crests.
Downstream DOCSIS 3.0 signals, on the other hand, may have a bandwidth of 6 MHz, and utilize quadrature amplitude modulation (QAM) schemes such as 64 QAM or 256 QAM. At such relatively narrower bandwidths, such modulation schemes exhibit lower RF power relative to the total power being amplified, and similarly have a lower crest factor relative to the OFDM signals. One conventional diagnostic technique for downstream signals is a “truck-roll,” which measures nonlinear distortion with equipment which physically moves along the signal path, and measures distortion on the signals. Diagnostic techniques presently exist, but typically require that an in-service carrier be taken out of service so that a test signal may be transmitted along the signal path. Exemplary systems and methods for measuring nonlinear distortion in the vacant band are described in greater detail in U.S. Pat. Nos. 9,209,863, 9,225,387, and 9,590,696, the disclosures of which are incorporated by reference herein.
Additionally, conventional diagnostic techniques for nonlinear distortion in downstream OFDM signals do not tend to produce significant or meaningful results. For example, where a 6 MHz channel is transmitted as just one of as many as 150 other channels, diagnostic test results of the one 6 MHz channel will not stand out significantly from the other channels. That is, where a square constellation is produced in the time domain, the test results from a saturated amplifier may simply appear as merely corners of the square constellation being pushed inward towards the origin. Thus, when measuring a single carrier out of 150 carriers, the corners of the resulting constellation would not exhibit significant compression, since the distortion from other uncorrelated carriers would dominate. In contrast, in the case of single carrier transmissions (e.g., wireless), corner compression would be more apparent. Accordingly, it is further desirable to be able to remotely perform nonlinear distortion diagnostic testing without interrupting service on an in-service carrier.